[I]t would be absurd to deny that in characterizing various actions as mercenary, cruel, or deceitful, sociologists are frequently (although perhaps not always wittingly) asserting appraising as well as characterizing value judgments. Terms like ‘mercenary' ‘cruel' or ‘deceitful’ as commonly used have a widely recognized pejorative overtone. Accordingly, anyone who employs such terms to characterize human behavior can normally be assumed to be stating his disapprobation of that behavior (or his approbation, should he use terms like ‘nonmercenary' ‘kindly' or ‘truthful’), and not simply characterizing it.
However, although many (but certainly not all) ostensibly characterizing statements asserted by social scientists undoubtedly express commitments to various (not always compatible) values, a number of “purely descriptive" terms as used by natural scientists in certain contexts sometimes also have an unmistakably appraising value connotation. Thus, the claim that a social scientist is making appraising value judgments when he characterizes respondents to questionnaires as uninformed, deceitful, or irrational can be matched by the equally sound claim that a physicist is also making such judgments when he describes a particular chronometer as inaccurate, a pump as inefficient, or a supporting platform as unstable. Like the social scientist in this example, the physicist is characterizing certain objects in his field of research; but, also like the social scientist, he is in addition expressing his disapproval of the characteristics he is ascribing to those objects.
Nevertheless—and this is the main burden of the present discussion—there are no good reasons for thinking that it is inherently impossible to distinguish between the characterizing and the appraising judgments implicit in many statements, whether the statements are asserted by students of human affairs or by natural scientists. To be sure, it is not always easy to make the distinction formally explicit in the social sciences—in part because much of the language employed in them is very vague, in part because appraising judgments that may be implicit in a statement tend to be overlooked by us when they are judgments to which we are actually committed though without being aware of our commitments. Nor is it always useful or convenient to perform this task. For many statements implicitly containing both characterizing and appraising evaluations are sometimes sufficiently clear without being reformulated in the manner required by the task; and the reformulations would frequently be too unwieldy for effective communication between members of a large and unequally prepared group of students. But these are essentially practical rather than theoretical problems. The difficulties they raise provide no compelling reasons for the claim that an ethically neutral social science is inherently impossible.-- Ernest Nagel (1961) The Structure of Science: Problems in the Logic of Scientific Explanation, pp. 494-495 (emphasis in original)
I noted the other day that Ernest Nagel's attempt to distinguish between characterizing and appraising evaluations are explicitly directed against a series of arguments by Leo Strauss. From the point of view of the sociology of philosophy the passage is remarkable because it is a rare case where Strauss is treated as a serious interlocuter by an analytic philosopher and where much of Strauss' position is conceded, and the remainder of the differences met by an argument.
As regular readers know, I am a fan of Nagel. But the section is not his finest. First, it's not mentioned by Nagel that Strauss' arguments are an immanent critique of Max Weber. And, as I noted, the way Strauss' arguments are quoted this is obscured. Second, in arguing against Strauss, Nagel makes it seem that Strauss is a critic of objectivity in science. In particular, Nagel makes it seem that the critic of value neutrality in social science must be a critic of the value neutrality of natural science and de facto committed to what we now call the 'strong program.'* Whereas Strauss' position is that objectivity in social science is constituted by a commitment to the value-laden-ness of social science.
Now, as it happens, and I as I also noted, in a recent and already much cited article, Anna Alexandrova has written critically on Nagel's position (and notes explicitly that it was directed against Strauss). And in what follows I draw on her position. But before I get to that it is also worth noting that there exists a published transcript of Strauss's course, “Introduction to Political Philosophy” offered in the winter term of 1965, open to undergraduate as well as graduate students. It was published as Leo Strauss on Political Philosophy: Responding to the Challenge of Positivism and Historicism Edited by Catherine H. Zuckert (The University of Chicago Press, 2018). In it, Strauss responds to Nagel's criticism and also develops his own criticism of Nagel, too, in quite interesting fashion.+
Let's turn to Alexandrova. She introduces special terminology in order to characterize the claims of interest in the debate between Nagel and Strauss (and, in fact, in the debate about the value-ladenness of social science). I quote her:
- A hypothesis is mixed if and only if:
- 1. It is an empirical hypothesis about a putative causal or statistical relation.
- 2. At least one of the variables in this hypothesis is defined in a way that presupposes a moral, prudential, political or aesthetic value judgment about the nature of this variable.
The key part here is (2). In fact (1) is a place-holder or short-hand for more expansive claims about empirical hypotheses. And one should not be misled by the use of 'hypothesis.' It's okay for Alexandrova's position if one treats a so-called mixed claim as established or confirmed. It's, however, important to recognize that the values in (2) are not the so-called "cognitive values – simplicity, explanatory power, coherence, generality etc. –" but the "non-cognitive values – moral, prudential, political or aesthetic –....(Longino [1990], Lacey [2005])." (Alexandrova). It is the second kind that figures in mixed claims.
Alexandrova usefully distinguishes mixed claims from other kinds of value-ladenness in science. But her position is -- and here she agrees with Strauss -- that in many contexts, mixed claims cannot be eliminated from social science (although she notes that often their presence goes unnoticed), but that this is not itself a barrier to an objective social science. (To be sure, there are real differences between Alexandrova's position and Strauss, but those I leave aside here today.)
Okay, ley's now return to Nagel's distinction between characterizing and the appraising judgments. As the passage above makes clear, Nagel allows that in practice scientists may well conflate the two or may well use terms that are treated as descriptive, but that in reality also have some kind of appraisal built into them. (One may add that philosophers often introduce special terminology with the same character.) But he asserts that it is not "inherently impossible to distinguish between the characterizing and the appraising judgments implicit in many statements." (emphasis in original)
Alexandrova herself suggests that "A natural way to implement Nagel’s proposal is to convert mixed claims from regular causal or correlational claims into conditional claims." But as Alexandrova convincingly shows, if that's Nagel’s approach then it "eliminates values at one stage, but it only pushes them to another less appropriate stage." Alexandrova then works through three ways in which converting mixed claims into conditional claims does not work. I found it convincing, but go read her paper!
Interestingly enough, in the quoted passage above, Nagel suggests that there is a vantage point from which one can evaluate and "reformulate" statements used in science and then distinguish between their characterizing and appraising content. In fact, this vantage point is occupied by the Nagelian philosopher of science. This is not obvious from context, but it is the stated aim of Nagel's Structure. As Nagel puts it in his "Introduction:"
However, if the nature of the scientific enterprise and its place in contemporary society are to be properly understood, the types and the articulation of scientific statements, as well as the logic by which scientific conclusions are established, also require careful analysis. This is a task—a major if not exclusive task—that the philosophy of science undertakes to execute. (p. 14)
That is to say, Nagel here comes very close to the idea associated with Quine that one can regiment science and that it is the philosopher of science's task to do so. Today I leave open to what degree Quine or Nagel influenced each other (and questions of priority).
Nagel doesn't say how one should regiment 'mixed claims.' He only claims that for practical purposes it can be done.** That's to say, he thinks there is a vantage point from which appraising value connotations in apparent mixed claims can be kept at arm's length. (This is why Alexandrova's suggestion that he is thinking of turning them into conditional statements is so plausible.) But because he does not explain how that practical (ahh) stance can be achieved -- and why in regimentation we can be properly neutral--, he raises the suspicion that much of the rhetorical work of his argument is being done by the implication that if one insists that social science cannot eliminate mixed claims, then one must be an adherent to the 'strong program' or be a skeptic about the possibility of an objective social science.
*In broader context, Nagel is critical of Karl Mannheim, so the connection to the strong program is not accidental.
+I thank Jeffrey Bernstein
Comments